I finally created a vector graphic version of the 128 Icelandic cardinal directions (see post from March 18 for the original version). Created in AutoCAD 2015.
One of the more interesting variables in the Citi Bike System data is gender, although it’s only recorded if the user is registered. Turns out that the majority of the Citi Bike users are males…why? And why are male trips constantly shorter in minutes than female trips? Are men just faster? The gender variable has been mapped before but this is my version of it. This is the first time I create a map with the chart option in ArcGIS and I´m not impressed. I usually rotate Manhattan so the avenues are vertical (the magic number is 28.5° CCL) but this will also rotate the pie labels…and there is no way to disable that relationship. Note that I left some Citi Bike stations out because of clutter. So I left Manhattan unturned.
The Icelandic Student Loan Fund supports students with loans at educational institutions both in Iceland and abroad. This covers both tuition and cost of living and has given thousands of Icelandic students the opportunity to seek higher education around the world (author included). Recently the Fund quite abruptly decided to decrease the basic support rate for all foreign countries by as much as 10% for the 2014-2015 period (two semesters in most countries), claiming it was too high and hinting that for some cases more than 10% decrease would be justifiable . For each period the Fund issues allocation rules where the basic support rate is compiled for a shortlist of countries. This map shows the change in support rates from the previous period (2013-2014) to the current period (2014-2015). Since the Fund changes the number of countries each time the map only shows countries that are comparable between the two periods (50).
Statistically significant hot spots (and cold spots) for NYC Bike Share Program (Citi Bike) with two distance bands, 500 meters on the left and 1,000 meters on the right….decisions, decisions, decisions. Hot Spots generated using the Geits-Ord Gi* statistic for system data counts (bike picked up or returned). The 500 m will include 8 neighbors on average but the 1,000 m will always include at least 8 neighbors. Part of a term project at Lehman College course GEP 630 with Prof. Johnson (Geo-statistics).
Lehman College course GEP 630 with Prof. Johnson (Geo-statistics). Bicycle collisions from August 2011 till February 2014 in Manhattan, sub 59th street mapped with intersection catchment areas. I’m taking my bicycle collision study a bit further this semester; I’ll be bringing this data into SatScan for some spatio-temporal cluster analysis which does not require any population at risk (No one really know how many cyclists there are in NYC). Hoping to push this even further and do some spatial regression analysis and trying to predict the collisions with other variables like Citibike usage data and if there are bike lanes or not. Since the collisions are all aggregated to intersections it made sense to me to create intersection catchment areas or catchment streets. (Collision geocoded by The NYPD Crash Data Band-Aid).
— George E. P. Box
Cardinal directions are one of the basic elements of cartography. I’ve had this scan for a while. It names all the 128 cardinal directions in the Icelandic language. The original belonged to my grandfather and hangs in the family summerhouse back in Iceland. It seems that finally I might have to digitize this and create a proper vector graphic version. Perhaps this will be my contribution to safeguard the Icelandic language maybe. But having 128 words to choose from when pointing at something is far from practical, but it’s cool right?
Lehman College course GEP 630 with Prof. Yuri Gorokhovich (Natural Hazards and Disasters). The goal of this nifty little exercise was to estimate the impact of a 10 meter Tsunami near JFK airport. No instructions given, just data.
Lehman College course GEP 630 with Prof. Johnson (Geo-statistics). The 3 maps show rates mapped to the to the ZIP code level. The first one uses the raw rates but the second and the third use two different types of parametric smoothing called Empirical Bayes (EB), a global version and a local version. Note that although all the maps show rates classified into 5 classes (natural breaks) the values for each class are not the same when the maps are compared.